VMware 2V0-13.25 - VMware Cloud Foundation 9.0 Architect

Logical design decisions define how conceptual requirements are realized by technology features and configurations.

A is a logical design choice because it specifies how replication will be configured (synchronous) to meet an RPO requirement.

B is a requirement, not a design decision.

C is a physical design choice, as it specifies subnetting (/16 networks).

D is an application-level requirement, not a VCF infrastructure logical decision.

Thus, the logical design decision is using synchronous replication to satisfy RPO.

A single VCF instance with dedicated Workload Domains per tenant strikes the balance between operational efficiency and isolation. It reduces lifecycle tasks since only one management domain must be maintained, while each tenant having a dedicated workload domain ensures isolation of compute resources. This meets all three stated requirements effectively: lifecycle simplicity, minimal overhead, and tenant-specific compute separation.

The VMware Cloud Foundation 9.0.1 Architecture Overview and Operations Integration Guide state that VCF Operations, combined with VCF Operations for Logs, provides a unified monitoring and observability solution across the entire VCF fleet. The documentation specifies:

“Integrating VCF Operations with VCF Operations for Logs and VCF Automation provides a single-pane-of-glass for monitoring metrics, events, and flows across all management and workload domains.”

Furthermore, the High Availability deployment model for VCF Operations for Logs is required to meet the 99.9% SLA for continuous availability of monitoring data.

“VCF Operations for Logs can be deployed in High Availability mode to ensure log continuity and fault tolerance.”

Thus, the architect must:

Integrate VCF Operations with both Logs and Automation (B, D) for unified visibility.

Deploy Logs in High Availability (F) mode to achieve the 99.9% availability SLA.

References (VMware Cloud Foundation documents):

VMware Cloud Foundation 9.0.1 Operations Design Guide — “Integrated Operations and Log Monitoring Architecture.”

VMware Cloud Foundation 9.0.4 Integration Guide — “VCF Operations and Logs HA Deployment.”

According to the VMware Cloud Foundation 9.0.4 Architecture and Design Guide, the officially supported solution for backing up and restoring workloads running on VMware Kubernetes Service (VKS) clusters is Velero, specifically the Velero Plugin for vSphere. The documentation explicitly states:

“You can backup and restore workloads running on VKS clusters by using Velero. The Velero Plugin for vSphere provides a solution for backing up and restoring workloads running on vSphere Pods and VKS clusters.”

Velero integrates with vSphere through a plugin that manages persistent volume snapshots and metadata backups to an S3-compatible object store. This provides data protection for both stateless and stateful Kubernetes workloads.

While Restic is mentioned in conjunction with Velero for additional support in some environments, it cannot operate standalone in VKS environments. Site Recovery Manager and VMware Live Recovery are used for virtual machine–based disaster recovery, not containerized workloads. Therefore, Velero is the correct and VMware-supported tool for VKS workload backup and recovery.

References (VMware Cloud Foundation documents):

VMware Cloud Foundation 9.0.4 — “Backing Up and Restoring VKS Clusters and Workloads” (pp. 5879–5887)

VMware Cloud Foundation 9.0.4 — “Install and Configure the Velero Plugin for vSphere on a Supervisor” (pp. 5880–5885)

Due to the explicit constraint that no internet access is permitted, the VCF Installer cannot connect to an online depot. Instead, the architect must use the offline depot model, where binaries are downloaded externally and made accessible locally within the VCF environment (e.g., using a local web server).

This setup aligns with VMware's "air-gapped" deployment guidance for VCF environments with strict security postures.

According to the VMware Cloud Foundation 9.0.4 Architecture and Design Guide, vSphere Pods (which support VMware Kubernetes Service or VKS clusters) use three types of storage: ephemeral VMDKs, persistent volume VMDKs, and container image VMDKs.

The document explicitly states:

> “A vSphere Pod requires ephemeral storage to store such Kubernetes objects as logs, emptyDir volumes, and ConfigMaps during its operations. This ephemeral, or transient, storage lasts as long as the pod continues to exist.”

Ephemeral storage is non-persistent by design and is deleted once the pod lifecycle ends. It provides temporary space for data that doesn’t need to persist beyond the pod’s lifespan, such as application logs, temporary caches, or transient compute data. This aligns with Kubernetes’ ephemeral storage model integrated into vSphere infrastructure.

Persistent workloads, by contrast, utilize storage policies and vSphere CNS-backed persistent volumes. Therefore, ephemeral storage is the correct type for non-persistent pod data.

References (VMware Cloud Foundation documents):

VMware Cloud Foundation 9.0.4 — “Storage Policies for a Supervisor” (p. 5632–5633)

VMware Cloud Foundation 9.0.4 — “vSphere Pod Storage Types: Ephemeral, Persistent, and Container Image VMDKs.”

---

Since the constraint is existing stretched cluster model, the design assumes a single cluster across two sites.

To meet the minimize downtime requirement, the Single Management Zone Supervisor with HA control plane ensures control plane resiliency within that stretched cluster.

Options with “Simple Availability” do not meet the high availability requirement. Options with three zones are not applicable because the constraint specifies a stretched cluster model.

In VMware Cloud Foundation, each VI workload domain is backed by a single vCenter Server instance. By deploying multiple clusters within the same VI workload domain, the architect can support multiple use cases (e.g., separating prod/dev), without provisioning new vCenters. This design reduces management overhead and operational complexity.

However, if stricter separation is needed (e.g., multi-tenancy or lifecycle independence), separate workload domains may be more suitable. While vSAN is the default, it's not mandatory unless vSAN Ready Nodes are used for bring-up.

RAID-1 policies offer low latency and high performance, ideal for critical workloads such as transactional databases.

RAID-5/6 provides efficient capacity usage, which is preferred for non-critical workloads.

VMware recommends the vSAN Express Storage Architecture (ESA) with 6 nodes minimum for performance and redundancy.The ESA supports better performance with fewer overheads than OSA. VMware Cloud Foundation 9.0 promotes using vSAN ESA for mission-critical workloads when latency is a priority.

The VMware Cloud Foundation 9.0.1 Architecture Guide defines three levels of design abstraction — Conceptual, Logical, and Physical. The Physical Design translates logical components into tangible configuration and implementation details. VMware describes it as:

“The physical design includes the specific details for hardware models, network topologies, storage layouts, port configurations, VLAN IDs, and zoning of Fibre Channel fabrics.”

In contrast:

The Conceptual Design defines what the solution must deliver (high-level goals and relationships).

The Logical Design outlines component relationships and service flows without vendor-specific or configuration details.

Therefore, the Physical Design is where the architect defines server model types, port mappings, uplink configurations, vSAN disk group layouts, and Fibre Channel zones, all aligned to the validated VCF Bill of Materials (BOM).

References (VMware Cloud Foundation documents):

VMware Cloud Foundation 9.0.1 Design Guide — Conceptual, Logical, and Physical Design Definitions (pp. 79–81).

VMware Cloud Foundation 9.0.2 Architecture Overview — Physical Design Implementation Detailing Hardware and Network Configuration.